Search

INTRODUCTION: Post‐mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life. METHODS: This study employed statistical tools and machine learning to predict 17 neuropathologic lesions from a cohort of 6518 individuals using 381 clinical features (Table S1). The multisite data allowed validation of the model's robustness by splitting train/test sets by clinical sites. A similar study was performed for predicting Alzheimer's disease (AD) neuropathologic change without specific comorbidities. RESULTS: Prediction results show high performance for certain lesions that match or exceed that of research annotation. Neurodegenerative comorbidities in addition to AD neuropathologic change resulted in compounded, but disproportionate, effects across cognitive domains as the comorbidity number increased. DISCUSSION: Certain clinical features could be strongly associated with multiple neurodegenerative diseases, others were lesion‐specific, and some were divergent between lesions. Our approach could benefit clinical research, and genetic and biomarker research by enriching cohorts for desired lesions. [ABSTRACT FROM AUTHOR]

Background: Neuropathologic changes are central for both understanding of the patients' brains and making the definitive diagnosis of dementia‐related diseases. However, many of them are only obtainable post‐mortem. To make this information available while an individual is still alive, this study developed machine learning models that predict neuropathologic changes based on features obtainable during life. Method: The multi‐site post‐mortem data (n∼5000) was obtained from National Alzheimer's Coordinating Center. A multitask long‐short term memory‐based neural network architecture was developed with custom loss to predict the 13 neuropathologic changes from features during life measured longitudinally. The performance of the model was evaluated using entire unseen sites as test sets. Evaluation metrics include area under receiver's operating curve (AUROC) and area under precision recall curve (AUPRC). Result: The model was able to predict Alzheimer's Disease neuropathologic changes (ADNC) and any Alzheimer's‐related pathologies, such as Braak score, with great sensitivity, specificity, and precision (for example AUROC = 0.85; AUPRC = 0.96 for ADNC). Apart from these, the model can also predict hippocampal sclerosis accurately (AUROC = 0.79; AUPRC = 0.80) and Lewy Body disease at higher precision than clinician's diagnosis. Model interpretation shows patterns in neuropsychological tests that are predictive of pathologic changes. Additionally, model error analysis revealed factors, such as resilient case ratios, that explain variation in performance between sites. Conclusion: Patterns of measurable features during life can be used by machine learning to predict ADNC, hippocampal sclerosis, and to lesser extent Lewy Body. [ABSTRACT FROM AUTHOR]

Background: Cerebrospinal fluid (CSF) neurofilament light protein (NfL) is a sensitive biomarker of central nervous system neuro‐axonal injury. A novel assay measuring blood NfL has been developed, avoiding the need for CSF sampling. Several studies have published correlations between CSF and blood NfL, but the pooled correlation is unknown. This meta‐analysis aims to determine the pooled correlation coefficient estimate between CSF and blood NfL, and to stratify by correlation technique and blood NfL assay. Method: We searched Medline, Embase and Web of Science for published manuscripts in English, from their inception to July 9, 2019, according to PRISMA guidelines. We included studies reporting the correlation between CSF and blood (plasma/serum) NfL. We then conducted a random‐effects meta‐analysis to calculate the pooled correlation coefficient estimate, accounting for correlation technique and assay used. Heterogeneity was assessed using the Cochran Q test. In sensitivity analyses, we calculated the pooled correlation coefficient estimate according to whether Spearman's rank or Pearson's correlation coefficient was reported (where unspecified, Pearson's correlation was imputed), and according to blood NfL assay: single‐molecule array digital immunoassay (Simoa), electrochemiluminescence (ECL) or enzyme‐linked immunosorbent assay (ELISA). Result: Data were extracted from 37 articles, including 3797 individuals. 23 studies reported Spearman's rank and 10 studies reported Pearson's correlation coefficient; correlation technique was unspecified in 4 studies. 29 studies used Simoa, 7 ECL and 1 ELISA. Results from the overall meta‐analysis demonstrated that the pooled correlation coefficient estimate for CSF and blood NfL was r=0.72, 95% confidence interval 0.56‐0.83, p<0.001 (Figure 1). Heterogeneity was significant: Q=206.2, p<0.0001. In sensitivity analyses, for studies quoting Spearman's rank, the pooled estimate was r=0.66, p<0.001, while the pooled estimate was r=0.76, p<0.001 for studies quoting Pearson's correlation. When stratified by blood NfL assay, the pooled correlation coefficient was similar for Simoa and ECL (r=0.70, p<0.001 and r=0.73, p<0.001, respectively), but weaker for ELISA (r=0.35, p=0.272). Conclusion: Strong correlations are demonstrated between CSF and blood NfL, especially with Simoa and ECL blood NfL assays. Simoa is the preferred blood NfL assay due to its low detection limit, and this meta‐analysis supports blood NfL measured using Simoa, as a surrogate measure of CSF NfL. [ABSTRACT FROM AUTHOR]